JP2010198166A - Periphery monitoring device for vehicle - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a periphery monitoring device of a vehicle for executing processing for determining an object in the periphery of a vehicle as necessary. <P>SOLUTION: A vehicle periphery monitoring device for monitoring the periphery of a vehicle by using a pickup image to be obtained by an image pickup means loaded on a vehicle includes: a means for detecting a location where a vehicle is traveling; a determination means for determining whether or not appearance possibility information showing possibility that an object to be detected appears is set about the detected location; and a start means for, when the appearance possibility information is set, starting processing for determining whether or not the object extracted from the pickup image is an object to be detected. The object to be detected is, for example, an animal, and animal determination processing is executed only while the vehicle travels in the location where there is possibility that an animal appears, so as to reduce the processing load. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an apparatus for monitoring the periphery of a vehicle, and more specifically, to an apparatus that executes a process of determining an object around the vehicle as necessary.

  In the vicinity of the vehicle, there are pedestrians, artificial structures such as utility poles and walls, etc., and such objects may affect the running of the vehicle. There has been proposed an apparatus that is mounted and detects an object from captured images around the vehicle imaged by the camera and provides the detection result to the driver of the vehicle. In addition, there are cases where animals such as deer are present around the vehicle, and recently, a method for detecting such animals separately from other objects has been proposed. For example, Patent Document 1 below discloses a technique for detecting an animal. According to this technique, an animal is detected by extracting an object having a predetermined shape representing the animal from a captured image.

JP 2007-310705 A

  Living bodies such as pedestrians and animals are usually at a higher temperature than artificial structures, and therefore, using an infrared camera, these living bodies can be detected from the artificial structures by distinguishing them to some extent. However, in order to distinguish and determine these objects with higher accuracy, a separate object determination algorithm is required. In particular, since a pedestrian and an animal are the same living body, separate object determination algorithms are required to distinguish and detect both. There are various types of animals such as deer, bears and horses. In addition, there are not one type of deer, but there are various types of deer that vary in size. The same applies to bears and horses. If these various animals are to be detected without omission, the number of object determination algorithms may increase, or the contents of the algorithms may become complicated. As the number of algorithms provided for each type of object to be detected increases and the algorithm becomes more complicated, the processing load increases and the processing efficiency may decrease.

  Therefore, there is a demand for an apparatus that can suppress an increase in the processing load as described above and can determine an object to be detected more efficiently.

  According to one aspect of the present invention, a vehicle periphery monitoring device that monitors the periphery of a vehicle using a captured image obtained by an image capturing unit mounted on the vehicle includes: a unit that detects a position where the vehicle is traveling; Determination means for determining whether or not the presence / absence possibility information indicating the possibility that the object to be detected appears or not is set for the position that has been detected; And an activation unit that activates a determination process for determining whether the object is the object to be detected.

  According to the present invention, the object determination process is not always performed, but is performed only when the vehicle exists at a position where the object may appear and disappear, so that the processing load is reduced. Can be achieved. In addition, since the object determination process is performed at a position where the object to be detected may appear and disappear, the object determination accuracy can be further improved.

  According to an embodiment of the present invention, the object to be detected is an animal, and the appearance possibility information is set for a predetermined section including a position indicating the possibility that the animal appears. The determination means determines whether or not the detected position is included in the predetermined section. Only when it is determined that the detected position is included in the predetermined section, the activation unit activates a determination process for determining whether or not the object extracted from the captured image is an animal.

  You can predict to some extent where the animals will appear. For example, the area (predetermined section) where the animal appears can be predicted from the marker information about the animal and the information that the animal has appeared in the past. The present invention has been made in view of this finding, and the animal determination process is executed only when the vehicle is traveling in a section where such information exists. Since the determination process is executed only in an area where an animal is likely to appear, the processing load can be reduced, erroneous detection can be suppressed, and the animal determination accuracy can be improved.

  According to an embodiment of the present invention, the haunting possibility information includes at least one of a kind of an animal that may appear and an animal size for each kind, and the activation means includes the The determination process to be activated is selected or changed according to at least one of the type of animal included in the appearance possibility information set for the detected position and the size of the animal for each type.

  There are various types of animals, and the size may be different for each type. According to the present invention, it is possible to select or change the animal determination process so as to be adapted to an animal of a kind or size that may appear and disappear. For example, if the information that may appear and disappear is a deer, the animal determination process for a deer can be started, and if it is a bear, the animal determination process for a bear can be started. In addition, even in the same deer, the contents of the animal determination process (for example, a threshold for determining the size) can be changed so as to match the size of a deer that may appear and disappear. Since only the determination process for animals of a limited kind and size needs to be activated, the processing load can be reduced. In addition, since the determination process is performed only for an animal of a type or size that may appear and disappear, erroneous detection can be suppressed, and the determination accuracy of the animal can be improved.

  Other features and advantages of the present invention will be apparent from the detailed description that follows.

The block diagram which shows the structure of the periphery monitoring apparatus according to one Example of this invention. The figure for demonstrating the attachment position of the camera according to one Example of this invention. The figure which shows the appearance possibility information according to 1st-3rd Example of this invention. The figure which shows the example of the label | marker regarding an animal. The figure which shows that the magnitude | size of an animal changes according to the kind of animal. 3 is a flowchart illustrating a process in an image processing unit according to an embodiment of the present invention. The flowchart which shows the process which judges the necessity of the animal determination process according to 1st Example of this invention. The flowchart which shows the process which judges the necessity of the animal determination process according to 2nd Example of this invention. The flowchart which shows the process which judges the necessity of the animal determination process according to 3rd Example of this invention.

  Next, an embodiment of the present invention will be described with reference to the drawings.

  FIG. 1 is a block diagram showing a configuration of a vehicle periphery monitoring device according to an embodiment of the present invention. The apparatus is mounted on a vehicle and has two infrared cameras 1R and 1L capable of detecting far infrared rays, and an image processing unit 2 for detecting an object around the vehicle based on image data obtained by the cameras 1R and 1L. And a speaker 3 that generates an alarm by voice based on the detection result, and an image obtained by the camera 1R or 1L, and a head-up display that displays a display for allowing the driver to recognize an object around the vehicle (Hereinafter referred to as HUD) 4.

  In this embodiment, as shown in FIG. 2, the cameras 1 </ b> R and 1 </ b> L are arranged at positions symmetric with respect to the central axis passing through the center of the vehicle width at the front of the vehicle 10 so as to capture the front of the vehicle 10. Has been. The two cameras 1R and 1L are fixed to the vehicle so that their optical axes are parallel to each other and their height from the road surface is equal. The infrared cameras 1R and 1L have a characteristic that the level of the output signal becomes higher (that is, the luminance in the captured image becomes higher) as the temperature of the object is higher.

  The image processing unit 2 includes an A / D conversion circuit that converts an input analog signal into a digital signal, an image memory that stores a digitized image signal, a central processing unit (CPU) that performs various arithmetic processing, and a data RAM (Random Access Memory) used to store data, ROM (Read Only Memory) that stores programs executed by the CPU and data used (including tables and maps), driving signals for the speaker 3, display signals for the HUD 4, etc. Is provided. The output signals of the cameras 1R and 1L are converted into digital signals and input to the CPU. As shown in FIG. 2, the HUD 4 is provided so that a screen 4 a is displayed at a front position of the driver on the front window of the vehicle 10. Thus, the driver can visually recognize the screen displayed on the HUD 4.

  Further, the navigation unit 5 is connected to the image processing unit 2. The navigation unit 5 receives a GPS signal for measuring the position of the vehicle 10 using, for example, an artificial satellite, and detects the current position of the vehicle 10 based on the GPS signal. The navigation unit 5 includes a map information storage unit 7 that stores map information. The map information storage unit 7 can be realized in any storage device (such as a recording medium or a hard disk drive). The navigation unit 5 can read the map information stored in the map information storage unit 7 and display it on the display screen of a predetermined display device (or HUD 4).

  In this embodiment, in the map information stored in the map information storage unit 7, information indicating that there is a possibility of appearing and appearing at a position where an object to be detected may appear and appear (hereinafter, possibility of appearing and appearing). Information) is set. In the following examples, the object to be detected is an animal. Animals often have limited places where they appear and are characterized by being easy to predict in advance. If a place where there is a possibility of appearing in advance is known in advance, the process for determining whether the object imaged by the camera is an animal may be performed only for that place. By doing so, the frequency of executing the animal determination process is reduced, so that the processing load can be reduced.

  Here, the animal possibility information will be described with reference to FIG. FIG. 3 (a) shows the hauntability information according to the first embodiment. In the map information, data indicating that there is a possibility of appearing is stored in association with the position where the animal is likely to appear, and in this embodiment, the appearance flag is used as the data. For example, each position A1, A2,. . . , A haunting flag having a value of 1 is set and stored, which is stored in these positions A1, A2,. . . Indicates that there is a possibility that some animal will appear. Therefore, by referring to the map information, it is possible to determine whether or not the current position of the vehicle is a position where an animal may appear or disappear.

  Information about whether or not there is a possibility of appearing can be obtained from any source. For example, as shown in FIG. 4, signs related to animals are installed on the road. FIG. 4 (a) shows that a deer can possibly invade, (b) shows that a cow can possibly invade, and (c) shows that a bear may appear. It is a sign indicating that. About the position where these signs are installed, an appearance flag can be set and memorize | stored in map information like Fig.3 (a).

  Further, information on the fact that the animal has appeared in the past, information on the accident that the vehicle collided with the animal, etc. are obtained from an arbitrary information source (for example, news from some organization, etc.), and FIG. As described above, it can be registered in advance in the map information.

  Further, the user may be able to register information as shown in FIG. For example, on the map information displayed on the HUD 4, the user designates a position where there is a possibility of appearing and the image processing unit 2 sets an appearing flag for the designated position and stores it in the map information. be able to.

  Preferably, as shown in FIG. 3 (a), each position A1, A2,. . . Is set to include each position. The predetermined section can be an arbitrary range. For example, the section B1 with respect to the position A1 can be a section of 1 km before and after the position A1 (total 2 kilometers) on the road where the position A1 exists. Alternatively, the section B1 may be set to have a radius a kilometer (a is 1 kilometer, for example) with the position A1 as the center. By setting the predetermined section, it is possible to set a section (area) where an animal may appear and appear with a margin.

  When such a predetermined section is set, each section B1, B2,. . . On the other hand, information indicating the possibility of appearance (in this embodiment, an appearance flag) is set and stored.

  FIG. 3B shows the possibility of appearance according to the second embodiment, and information indicating the type of animal is stored in the map information in association with each position as possibility information. For example, the type of animal set for the map information position A1 is a deer, which indicates that a deer may appear at the map information position A1. Since the shape of an animal differs depending on the type, by setting such an animal type, it is possible to accurately select an animal determination process to be activated at that position.

  Information on what kind of animals may be infested can be obtained from the sign information, the infestation information from some information source, etc., as described in FIG. 3 (a). Alternatively, it may be set by registration from the user. For example, in the case of FIG. 4A, since the sign represents “deer”, “deer” is registered as the animal type at the position where this sign is installed.

  In addition, each position A1, A2,. . . Instead of the predetermined sections B1, B2,. . . May be used.

  FIG. 3C shows the possibility of appearance according to the third embodiment. In addition to the type of animal, information indicating the size of the animal of that type is associated with each position as possibility information. Stored in map information. For example, the type of animal set for the position A1 of the map information is “deer”, and the size is 140 cm for the head and trunk length and 95 cm for the shoulder height. This indicates that a deer of this size may appear at the position A1.

  Here, referring to FIG. 5 for reference, the average size for various “deers” is shown. Even with the same type of “deer”, there are types of deer that are further subdivided according to the habitat (deer from sika deer to wabiti are deer), and these deers are different in size. It can be seen that the size of sika deer is registered at position A1 in FIG.

  In this way, not only what kind of animal should be determined by setting the map information about what kind of animal of what size and size may appear. , You can determine what size animal should be judged.

  Information on the size of animals that may appear is obtained from sign information, appearance information from some information source, etc., as described in FIGS. 3 (a) and 3 (b). It may also be set by registration from the user.

  In addition, each position A1, A2,. . . Instead of the predetermined sections B1, B2,. . . May be used.

  As described above, in this embodiment, the appearance possibility information shown in FIG. 3 is stored in advance in the map information of the map information storage unit 7 of the navigation unit 5. When the map information is updated through, for example, reading from a storage medium, the appearance possibility information may be updated at the same timing. In addition, when the navigation unit 5 is equipped with a communication function with a predetermined computer such as a server and the map information is updated with information distributed via communication, the appearance possibility information is also updated in the same manner. Can do. Alternatively, a table as shown in FIG. 3 may be stored in a storage device (a memory, a storage medium, or a hard disk) separately from the map information.

  Moreover, you may make it acquire the appearance possibility information in real time using the communication function of the navigation unit 5. For example, the vehicle communicates with a predetermined computer (server or the like), and habitability information is stored in the server. The vehicle transmits the current position of the vehicle to the server, and the server can notify the vehicle of the presence / absence possibility information for the current position. Alternatively, a system that provides traffic information to the vehicle such as an existing VICS system may be used, and the habitability information may be transmitted to the vehicle as part of the traffic information.

  Furthermore, you may combine storage of the possibility information in map information with acquisition of the possibility information in real time. For example, based on animal signs and past appearance information, map information may be stored in advance, and new appearance information that is not yet stored in map information may be acquired in real time via communication. Also good.

  FIG. 6 is a flowchart illustrating a process performed by the image processing unit 2 according to one embodiment of the present invention. The process is performed at predetermined time intervals.

  In steps S11 to S13, the output signals of the cameras 1R and 1L (that is, captured image data) are received as input, A / D converted, and stored in the image memory. The stored image data is a gray scale image including luminance information.

  In step S14, the right image captured by the camera 1R is used as a reference image (alternatively, the left image may be used as a reference image), and the image signal is binarized. Specifically, a process is performed in which a region brighter than the luminance threshold ITH determined in advance by simulation or the like is set to “1” (white) and a dark region is set to “0” (black). By this binarization process, an object higher than a predetermined temperature is extracted as a white region.

  In step S15, the binarized image data is converted into run-length data. Specifically, with respect to an area that has become white due to binarization, the coordinates of the start point (the leftmost pixel of each line) of the white area (referred to as a line) in each pixel row, and the start point to the end point (each The run length data is represented by the length (expressed by the number of pixels) up to the pixel on the right end of the line. For example, if the white region in the pixel row whose y coordinate is y1 is a line from (x1, y1) to (x3, y1), this line consists of three pixels, so (x1, y1, 3) This is represented by run-length data.

  In steps S16 and S17, the object is labeled and a process for extracting the object is performed. That is, of the lines converted into run length data, a line having a portion overlapping in the y direction is regarded as one object, and a label is given thereto. Thus, one or a plurality of objects are extracted.

  In addition to the above processing, for a plurality of past captured images, the same object is tracked (tracked), and the distance to the object is calculated based on the two captured images of the cameras 1R and 1L. You may make it determine whether a target object exists in the predetermined range around a vehicle, and whether there exists a possibility of entering the predetermined range. Details of these processes are described in, for example, JP-A-2001-006096.

  Steps S <b> 21 to S <b> 28 indicate processing for determining what is the target object thus extracted. In step S21, it is determined whether the object is an artificial structure. If it is determined that the object is not an artificial structure (No in step S21), it is determined in step S22 whether the object is a pedestrian. If it is determined that the user is a pedestrian (Yes in step S22), the target is determined to be an alarm target that is an alarm target (step S27). If it is determined that the object is an artificial structure (step S21 is Yes), and if it is determined that the object is not a pedestrian (step S22 is No), an animal determination process necessity determination process is performed in step S23. This will be described later with reference to FIGS.

  In step S24, the result of step S23 is examined, and if an animal determination process is necessary (Yes in step S24), the process proceeds to step S25 to execute an animal determination process for determining whether the object is an animal. In step S26, the result of the animal determination process is examined, and if it is determined that the object is an animal (Yes in step S26), it is determined that the object is an alarm object (step S27). If it is determined that the object is an alarm object, an alarm is output to the driver. The alarm can be output in any form, and the target object may be highlighted (for example, displayed by enclosing the target object in a frame) on the display screen 4 a of the HUD 4 or via the speaker 3. The driver may be informed that the object is present.

  On the other hand, if the animal determination process is not required (No in step S24), and if the result of the animal determination process is determined not to be an animal (No in step S26), the object is not an alarm object. Determination is made (step S28).

Here, the process of determining whether the object is an artificial structure, the process of determining whether it is a pedestrian, and the process of determining whether it is an animal can be realized by any appropriate technique. For example, the process for determining an artificial structure is described in Japanese Patent Application Laid-Open No. 2007-310705, Japanese Patent Application Laid-Open No. 2008-276787, and the like. For example, the process for determining whether a person is a pedestrian is described in Japanese Patent Application Laid-Open No. 2007-241740, Japanese Patent Application Laid-Open No. 2007-334751, and the like. For example, the process for determining whether or not an animal is described in JP 2007-310705 A, JP 2007-310706 A, and the like.
FIG. 7 shows the flow of the animal determination necessity determination process executed in step S23 according to the first embodiment. In the first embodiment, the appearance possibility information described with reference to FIG. Accordingly, in the map information, a haunting flag is set for a predetermined section in which some animal may haunt.

  In step S31, the current position of the vehicle is acquired via the navigation unit 5. Whether or not the current position is included in the predetermined section in which the appearance flag described with reference to FIG. 3A is set with reference to the map information stored in the map information storage unit 7 in step S32 Judging. If the current position is included in the predetermined section (Yes in step S32), it indicates that there is a possibility that some animal may appear and disappear at the current position, so an animal determination process is required in step S33. Is determined. If the current position of the vehicle is not within the predetermined section (No in step S32), it can be considered that there is almost no possibility that an animal will appear and disappear at the current position. ).

  As described with reference to FIG. 6, if it is determined that the animal determination process is necessary, the determination in step S <b> 24 is Yes. Therefore, the animal determination process is executed in step S <b> 25 to determine whether the object is an animal. judge. On the other hand, if it is determined that the animal determination process is unnecessary, the determination in step S24 is No, and the process proceeds to step S28 without executing the animal determination process.

  Thus, the animal determination process is activated only while the vehicle is traveling in a predetermined section set as a section where an animal may appear and appears, and the animal determination process is performed while the vehicle is traveling outside the predetermined section. It does not start. Therefore, since it is not necessary to always start the animal determination process, the processing load can be reduced. In addition, since the animal determination process is executed for the section where the animal may appear, if the animal appears, the animal can be detected more reliably.

  In this embodiment, the “predetermined section” described with reference to FIG. As described above, the “predetermined section” may not be set. In this case, in step S32, it is determined whether the appearance flag is set for the current position of the vehicle, and if the appearance flag is set. It is determined that the animal determination process is necessary, and if the appearance flag is not set, it is determined that the animal determination process is not necessary.

  The determination of the object described with reference to FIG. 6 is typically realized by the CPU of the image processing unit 2 (FIG. 1) executing an object determination program stored in advance in the memory. The Therefore, more specifically, “animal determination processing” indicates an algorithm for animal determination described in the program, but the implementation form of this algorithm in the program is arbitrary. That is, the animal determination algorithm may be implemented in a program different from other algorithms such as pedestrian determination, or may be implemented as a part of another program such as pedestrian determination. In either case, “execute the animal determination process” means that the animal determination algorithm part is executed, and “do not execute the animal determination process” means that the animal determination algorithm part is not executed. This is true for the following examples as well.

  FIG. 8 shows a flow of the animal determination necessity determination process executed in step S23 according to the second embodiment. In the second embodiment, the appearance possibility information described with reference to FIG. 3B is used. Therefore, in the map information, the type of the animal that appears is set for the position where the animal may appear. In the second embodiment, an animal determination process is prepared for each type of animal. For example, for “deer”, animal determination processing for deer (which is an algorithm as described above), for “bear”, animal determination processing for bear, and for “horse”, animal determination for horse Processing is prepared in advance, and is incorporated into a predetermined program and stored in a memory.

  In step S41, the current position of the vehicle is obtained via the navigation unit 5. In step S42, the map information stored in the map information storage unit 7 is referred to, and it is determined whether or not the information regarding the type of animal described with reference to FIG. If it is set (Yes in step S42), it indicates that there is a possibility that the set type of animal will appear at the current position. Therefore, in step S43, for the set type of animal. Select the prepared animal judgment process. For example, if the animal type “deer” is set for the current position, the animal determination process prepared for the deer is selected. Thereafter, in step S44, it is determined that an animal determination process is necessary. On the other hand, if information on the type of animal is not set for the current position in step S42, it can be considered that there is almost no possibility that an animal will appear at the current position. NO).

  As described with reference to FIG. 6, if it is determined that the animal determination process is necessary, the determination in step S <b> 24 is Yes, so that the animal determination process (algorithm) selected in step S <b> 43 in step S <b> 25. Only (therefore, the animal determination process (algorithm) of the non-selected type of animal is not executed), and it is determined whether or not the object is an animal. On the other hand, if it is determined that the animal determination process is unnecessary, the determination in step S24 is No, and the process proceeds to step S28 without executing the animal determination process.

  In FIG. 3B, when the “predetermined section” is set as described above, in step S42, the current position acquired in step S41 is a predetermined position in which information on the type of animal is set. It is determined whether or not it is within a section, and if it is determined that it is within the predetermined section, the process proceeds to step S43, and if it is determined that it is outside the predetermined section, the process proceeds to step S45.

  In addition, as described above, the “animal determination processing” is more specifically an algorithm for animal determination, so there are algorithms for deer, algorithms for bears, and the like. Is optional. That is, the deer algorithm may be realized as a program different from the bear algorithm, or the deer algorithm and the bear algorithm may be included in one program.

  The effect described with reference to FIG. 7 applies to the second embodiment as well. Further, the effect of this embodiment will be described. Conventionally, in order to determine whether or not an object extracted from a captured image is an animal, it is performed by examining whether or not a feature common to animals exists. It was. For example, in Japanese Patent Application Laid-Open No. 2007-310705, since the torso of a quadruped animal generally has an elliptical shape, it is examined whether or not this elliptical shape exists in the object.

  However, in order to determine an object with better accuracy, it is preferable to specify what animal the object is. Therefore, as described above, in this embodiment, an animal determination process is prepared for each type of animal, and each animal determination process may include processing logic for determining whether or not the animal is of that type. it can. Typically, since the shape of an animal is different for each type, a certain type of animal determination processing includes processing logic for checking whether the shape of the type of animal exists in the object. For example, in the deer animal determination process, an object having a shape corresponding to a deer is extracted from a captured image, and the extracted object is prepared in advance using known shape matching (pattern matching). Then, the similarity between the two is calculated by matching with a pattern imitating a deer, and if the similarity is high, it is determined that the object is a deer. In the animal determination process for a bear, an object having a size corresponding to a bear is extracted, and a pattern imitating a bear is used in shape matching. For example, Japanese Patent Laid-Open No. 2007-310706 describes that a deer is determined using such a method. In addition, depending on the type of animal, the moving speed may be different, so the animal determination process may be performed in consideration of the moving speed. For example, in a deer animal determination process, an object in a captured image can be tracked in time series, and it can be determined whether the moving speed is within a predetermined range set as the moving speed of “deer”.

  If animal determination processing of all types of animals prepared in advance is performed on the extracted target object, it is possible to determine which type of animal the target object is. However, in this method, as the number of animal types increases, the processing load increases and processing time increases. According to the second embodiment of the present invention, for each position (predetermined section), a kind of animal that may appear and disappear is predicted in advance, and therefore it is only necessary to perform animal determination for this kind of animal. . Therefore, the processing load can be reduced as compared with the conventional method.

  FIG. 9 shows a flow of the animal determination necessity determination process executed in step S23 according to the third embodiment. In the third embodiment, the appearance possibility information described with reference to FIG. 3C is used. Therefore, for the position where the animal may appear, the type and size of the animal that appears are set in the map information. In this embodiment, as will be described later, the size information is not necessarily set. In the third embodiment, as in the second embodiment, a separate animal determination process (algorithm) is prepared for each animal type.

  In step S51, the current position of the vehicle is acquired via the navigation unit 5. In step S52, the map information stored in the map information storage unit 7 is referred to, and it is determined whether or not the information regarding the type of animal described with reference to FIG. If it is not set (No in step S52), it can be considered that there is almost no possibility that an animal appears and disappears at the current position. Therefore, in step S56, it is determined that the animal determination process is not necessary. If it is set (Yes in step S52), it indicates that there is a possibility that an animal of that kind may appear at the current position.

  In step S53, it is determined whether or not size information is set for the current position. If the information on the size is set (Yes in step S53), the parameter in the animal determination process is changed in step S54 according to the information on the size. In step S55, it is determined that the animal determination process is necessary.

  The parameters include any parameters used to determine whether the object is an animal. For example, it may be determined whether the object extracted from the captured image has a predetermined size (set according to the distance from the vehicle to the object). Whether or not it has a “predetermined size” is checked using a threshold value, and the value of this threshold value can be changed according to the information on the set size. For example, in the animal determination process for deer, in order to determine whether or not the extracted object is a deer's torso, the extracted object is within a predetermined range (threshold) corresponding to the deer's head torso length. You can judge whether or not. In this case, if the current position of the vehicle is the position A1, the predetermined range is adjusted to information on the size of the position A1 (head torso length is 140 centimeters, shoulder height is 95 centimeters). If 150 cm is set to a predetermined range and the current position of the vehicle is position A2, according to the information about the size of position A2 (head torso length is 255 cm, shoulder height is 185 cm), for example 245 to 265 centimeters are set to a predetermined range. Thus, since the threshold value is appropriately set according to the size of an animal that may appear and disappear, the determination accuracy by the animal determination process can be improved.

  On the other hand, if the size information is not set in step S53 (No in step S53), the process proceeds to step S55 without changing the animal determination process (that is, skipping step S54). In this case, the threshold value in the animal determination process is set to a value (can be an initial value) that matches the size that the animal of that type can take. For example, in FIG. 5, the minimum deer head torso length (sika deer) is 140 centimeters, and the maximum (elk) is 255 centimeters, so it is determined whether it is a deer torso as described above. The predetermined range for can be, for example, 130-270 centimeters to include the minimum and maximum values.

  As described with reference to FIG. 6, if it is determined that the animal determination process is necessary, the determination in step S <b> 24 is Yes. Therefore, in step S <b> 25, the modified animal determination process is executed, and the object is detected. Determine if is an animal. On the other hand, if it is determined that the animal determination process is not necessary, the determination in step S24 is No, so the process proceeds to step S28 without executing the animal determination process.

  In FIG. 3C, when the predetermined section is set as described above, in step S52, the current position acquired in step S41 is within the predetermined section in which information on the type of animal is set. The process proceeds to step S53 if determined to be within the predetermined section, and proceeds to step S56 if determined to be outside the predetermined section.

  The effects described with reference to FIGS. 8 and 9 are similarly applied to the third embodiment. Furthermore, according to the third embodiment, since the animal determination process is changed to match the size of an animal that may appear and disappear, the accuracy of animal determination can be improved.

  In the third embodiment, an animal determination process is prepared for each type of animal. However, the present invention is not limited to this, and an animal determination process is prepared so as to be applicable to all animals. May be changed using information about the size. In this case, it is not always necessary to set information regarding the type of animal for each position of the map information, and information regarding the size of the animal may be stored in association with each other. Therefore, in step S52, it is determined whether or not information regarding the size of the animal is set for the current position. If it is set, the process proceeds to step S54, and if not set, the process proceeds to step S56.

  Arbitrary information may be combined among the appearance possibility information of the first to third embodiments. For example, by combining FIGS. 3A to 3C, the appearance flag is set for each of the predetermined sections (or positions), and information on the type and size of the animal is set as an option. While the vehicle is traveling in a predetermined section, if both the animal type and size are set as an option for the predetermined section, an animal determination process for the type of animal is performed according to the size. If only the information on the animal type is set as an option, the animal determination process for that type of animal is started. If only the appearance flag is set, the general animal determination process is executed. (This is a process for determining whether or not an animal, for example, the process described in Japanese Patent Application Laid-Open No. 2007-310705).

  In the above embodiment, the case where the object to be detected is an animal has been described. In general, an animal is often limited in the area where the animal appears, so the present invention is more effective in determining the animal. However, the object to be detected is not necessarily limited to animals. For example, as shown in FIG. 6, the pedestrian determination process is always executed. In addition to this pedestrian determination process, a process for determining a child with better accuracy is provided, for example, with a school zone sign. The present invention can also be applied to the case where only the predetermined section is executed.

  In addition, this invention is not restricted to said embodiment, Various deformation | transformation forms are possible. For example, in the above embodiment, an infrared camera is used as the imaging unit. However, for example, a visible camera that can detect only normal visible light may be used. However, by using an infrared camera, the object extraction process can be further simplified, and the calculation load can be reduced.

  In the above embodiment, as shown in FIG. 1, two infrared cameras are used. However, the present invention is applied to an image captured using one infrared camera or one visible camera. Is also applicable. In the case of one camera, the distance value to the object may be detected using, for example, a radar, or the distance from the position where the object is imaged to the object in the captured image of the camera is estimated. You may make it do.

  As described above, specific embodiments of the present invention have been described. However, the present invention is not limited to these embodiments.

1R, 1L infrared camera (imaging means)
2 Image processing unit 3 Speaker 4 Head-up display

Claims (3)

A vehicle periphery monitoring device that monitors the periphery of a vehicle using a captured image obtained by an imaging means mounted on the vehicle,
Means for detecting a position where the vehicle is traveling;
Determination means for determining whether or not the presence / absence possibility information indicating the possibility that the object to be detected appears and disappears is set for the detected position;
If the appearance possibility information is set, an activation unit that activates a determination process for determining whether an object extracted from the captured image is the object to be detected;
A vehicle periphery monitoring device comprising: The object to be detected is an animal, and the possibility information is set with respect to a predetermined section including a position indicating the possibility that the animal appears.
The determining means determines whether the detected position is included in the predetermined section;
The activation means determines whether or not the object extracted from the captured image is the animal only when the determination means determines that the detected position is included in the predetermined section. Start up,
The vehicle periphery monitoring apparatus according to claim 1. The habitability information includes at least one of the types of animals that may appear and the size of the animals for each type,
The activation means selects or determines the determination process to be activated according to at least one of an animal type included in the appearance possibility information set for the detected position and an animal size for each type. change,
The vehicle periphery monitoring apparatus according to claim 1 or 2.

Images